Bouzidi on curved moving boundaries

[vaillant]

Dear all,

In my case I have a rotating cylinder. I am currently using a regularized boundary and I am adding velocity thanks to “addVelocityBoundary” and “RotatingLinear3D”. It work fine but I would like to use the Bouzidi boundary in order to avoid the staircase shape. So I tried to define it with :

lattice.defineDynamics( superGeometry,3,&instances::getNoDynamics<T,DESCRIPTOR>() );
offBc.addVelocityBoundary( superGeometry,3,cylinder );
RotatingLinear3D<T> velocity(O, axe_cylinder, w, converter.getCharLatticeVelocity()/converter.getCharPhysVelocity());
offBc.defineU( superGeometry,3,velocity );

The problem is that the computation diverges after few time step.
However, if I reduced the velocity it work.

Could someone tell me what is wrong ?

Thanks in advance

[mathias]

Dear Vaillant,

what happens if you define a dynamics, e.g. the bulk dynmics instead?

Best Mathias

[vaillant]

Dear Mathias,

By defining a dynamics with :

lattice.defineDynamics(superGeometry, 3, &bulkDynamics);

the result is the same, the computation diverges.

Best regards

[mathias]

Ok, after how many steps is it diverging? Can you uploud or send me by email the output as well as a paraview vizualisation image of the velocity magnitute in a apropriate plane at the first timestep where uMax>0.3. It is very difficult to tell by means of the forum. Best Mathias

[Author]

Posts